中草药

基于多模型机器学习挖掘中药复方治疗脓毒症的“药-效”规律

PDF下载

中草药 | 数据挖掘与循证医学 2026,57(1): 214-222

收起

中草药 |数据挖掘与循证医学 2026 , 57 (1) : 214 -222

基于多模型机器学习挖掘中药复方治疗脓毒症的“药-效”规律

全屏

郑梦瑶^1,2,3,4, 刘青松^2,3,4, 王子晨^2,3,4, 谢双奕^2,3,4, 申晗^2,3,4, 王雨宁^2,3,4, 丁玲^2,3,4, 徐嘉悦^2,3,4, 金钊⁵, 王雯^2,3,4, 孙鑫^1,2,3,4

作者信息

通讯作者:

王雯

作者简介:

郑梦瑶: 郑梦瑶,硕士研究生,研究方向为急重症循证中药学研究。E-mail:2112342008@stu.gdpu.edu.cn 刘青松: 刘青松,博士研究生,研究方向为急重症循证中药学研究。E-mail:kliu_qs2025@stu.scu.edu.cn

Mining “medicine-efficacy” rule of traditional Chinese medicine formulas in treatment of sepsis based on multi-model machine learning

ZHENG Mengyao, LIU Qingsong, WANG Zichen, XIE Shuangyi, SHEN Han, WANG Yuning, DING Ling, XU Jiayue, JIN Zhao, WANG Wen, SUN Xin

Affiliations

doi: 10.7501/j.issn.0253-2670.2026.01.020

文章导航

摘要

收起

目的探索中药治疗脓毒症的“药-效”关联，识别与降低病死率、改善急性生理与慢性健康评分II（acute physiology and chronic health evaluation Ⅱ，APACHE Ⅱ）等关键临床结局相关的核心中药，为中药辅助治疗脓毒症提供循证参考。方法系统检索截至2024年9月的9个中英文数据库，纳入中药复方治疗脓毒症的临床研究，构建脓毒症中药复方数据库。采用十折交叉验证比较9种机器学习模型的预测性能，并运用沙普利加性解释（Shapley additive explanations，SHAP）方法解析各临床结局的最优模型，识别关键药物及其贡献方向。结果多层感知器在预测中医证候评分、有效率、炎症与免疫指标、生化指标、器官功能障碍评分及病死率性能最佳；逻辑回归在血气分析和胃肠功能与肠黏膜屏上表现最优；支持向量机在血常规与APACHE II评分上预测效能最佳。SHAP分析显示，地黄、枳实、黄芪、附子、黄芩、厚朴等中药与病死率、APACHE II评分、炎症指标和胃肠功能等结局中具有较高正向贡献，位于“药-效”网络的核心节点。结论通过多模型比较与可解释性分析，构建了以临床结局为导向的中药“药-效”关联网络，揭示了核心中药在脓毒症治疗中的潜在关键作用，为中医药精准辨证用药及中药新药研发提供数据支撑和循证依据。

关键词

脓毒症 / 中药复方 / 机器学习 / 沙普利加性解释 / “药-效”规律 / 数据挖掘 / 地黄 / 枳实 / 黄芪 / 附子 / 黄芩 / 厚朴

Abstract

收起

Objective To explore the “medicine-efficacy” associations of traditional Chinese medicine (TCM) in the treatment of sepsis, identify core herbsrelated to reducing mortality and improving key clinical outcomes such as the acute physiology and chronic health evaluation II (APACHE II) score, and provide evidence-based references for TCM-assisted treatment of sepsis. Methods A systematic search of Chinese and English databases was conducted up to September 2024 to collect clinical studies on TCM compound prescriptions for sepsis. A comprehensive sepsis TCM formula database was established. Nine machine learning algorithms were compared using ten-fold cross-validation, and the optimal model for each clinical outcome was interpreted through the Shapley additive explanations (SHAP) method to identify key herbs and their contribution directions. Results The multilayer perceptron showed the best performance in predicting TCM syndrome scores, overall effectiveness, inflammatory and immune indicators, biochemical parameters, organ dysfunction scores, and mortality; logistic regression performed best for blood gas analysis along with gastrointestinal function and intestinal mucosal barrier outcomes; and the support vector machine achieved optimal predictive performance for routine blood tests and APACHE II scores. SHAP analysis revealed that Dihuang (Rehmanniae Radix), Zhishi (Aurantii Fructus Immaturus), Huangqi (Astragali Radix), Fuzi (Aconiti Lateralis Radix Praeparata), Huangqin (Scutellariae Radix), and Houpo (Magnoliae Officinalis Cortex) had positive contributions across outcomes such as mortality, APACHE II score, inflammatory markers, and gastrointestinal function, forming the core nodes of the “medicine-efficacy” network. Conclusion This study established a clinically oriented “medicine-efficacy” association network through multi-model comparison and explainable machine learning analysis. The findings highlight the potential key roles of several core herbs in improving major clinical outcomes of sepsis, providing data support and evidence-based basis for precise syndrome differentiation and medication in TCM, as well as for the research and development of new TCMs.

Key words

sepsis / traditional Chinese medicine formulas / machine learning / SHAP / “medicine-efficacy” rule / data mining / Rehmanniae Radix / Aurantii Fructus Immaturus / Astragali Radix / Aconiti Lateralis Radix Praeparata / Scutellariae Radix / Magnoliae Officinalis Cortex

引用本文

郑梦瑶, 刘青松, 王子晨, 谢双奕, 申晗, 王雨宁, 丁玲, 徐嘉悦, 金钊, 王雯, 孙鑫. 基于多模型机器学习挖掘中药复方治疗脓毒症的“药-效”规律. 中草药, 2026 , 57 (1) : 214 -222 . DOI: 10.7501/j.issn.0253-2670.2026.01.020

ZHENG Mengyao, LIU Qingsong, WANG Zichen, XIE Shuangyi, SHEN Han, WANG Yuning, DING Ling, XU Jiayue, JIN Zhao, WANG Wen, SUN Xin. Mining “medicine-efficacy” rule of traditional Chinese medicine formulas in treatment of sepsis based on multi-model machine learning[J]. Chinese Traditional and Herbal Drugs, 2026 , 57 (1) : 214 -222 . DOI: 10.7501/j.issn.0253-2670.2026.01.020

基金

收起

国家杰出青年科学基金项目 (82225049); 国家自然科学基金面上项目 (82574870); 四川省中医药管理局中医药科研专项 (2024zd023); 四川省中医药管理局中医药科研专项 (25ZDAZX008); 四川大学华西医院学科卓越发展1·3·5工程项目 (ZYGD23004)

参考文献

收起

文献

收起

参考文献引证文献

排序方式：

Singer M, Deutschman C S, Seymour C W, et al. The third international consensus definitions for sepsis and septic shock(sepsis-3)[J]. JAMA, 2016, 315(8):801-810.
Rudd K E, Johnson S C, Agesa K M, et al. Global,regional, and national sepsis incidence and mortality,1990-2017:Analysis for the global burden of disease study[J]. Lancet, 2020, 395(10219):200-211.
Xie J F, Wang H L, Kang Y, et al. The epidemiology of sepsis in Chinese ICUs:A national cross-sectional survey[J]. Crit Care Med, 2020, 48(3):e209-e218.
Evans L, Rhodes A, Alhazzani W, et al. Surviving sepsis campaign:International guidelines for management of sepsis and septic shock 2021[J]. Intensive Care Med,2021, 47(11):1181-1247.
李志军,王东强,李银平,等.脓毒性休克中西医结合诊治专家共识[J]. 中华危重病急救医学, 2019(11):1317-1323.
Liu S Q, Yao C, Xie J F, et al. Effect of an herbal-based injection on 28-day mortality in patients with sepsis:The EXIT-SEP randomized clinical trial[J]. JAMA Intern Med,2023, 183(7):647-655.
张庆,姬文帅,孔欣欣,等.基于关联规则和隐结构模型的《普济方》中治疗喘证方剂的用药规律分析[J]. 中草药, 2023, 54(5):1517-1525.
Xia P, Gao K, Xie J D, et al. Data mining-based analysis of Chinese medicinal herb formulae in chronic kidney disease treatment[J]. Evid Based Complement Alternat Med, 2020, 2020:9719872.
吴文玉,詹少锋,焦欣,等.中医药领域机器学习研究的现状与发展趋势探讨[J]. 中华中医药学刊,(2025-07-30)[2025-12-25]. https://link.cnki.net/urlid/21.1546.R.20250730.1634.002.
姜皓,张冰,张晓朦,等.基于4种机器学习算法的妊娠期中药“禁忌慎”判别[J]. 中草药, 2021, 52(24):7596-7605.
油雨忻,李若轩,段梦雨,等.基于支持向量机的中药升降浮沉药性识别模型[J]. 时珍国医国药, 2022,33(11):2801-2804.
Chen H Y, He Y. Machine learning approaches in traditional Chinese medicine:A systematic review[J]. Am J Chin Med, 2022, 50(1):91-131.
中国药典[S]. 一部. 2020:1-6.
赵国平,戴慎,陈仁寿主编,南京中医药大学编著.中药大辞典[M]. 第2版.上海:上海科学技术出版社,2006:1-3091.
Yates Coley R, Liao Q Q, Simon N, et al. Empirical evaluation of internal validation methods for prediction in large-scale clinical data with rare-event outcomes:A case study in suicide risk prediction[J]. BMC Med Res Methodol, 2023, 23(1):33.
Lundberg S M, Erion G, Chen H, et al. From local explanations to global understanding with explainable AI for trees[J]. Nat Mach Intell, 2020, 2(1):56-67.
孔令博,晏军,张迪,等.中医药治疗脓毒症领域临床优势的探讨[J]. 中国实验方剂学杂志, 2022, 28(5):211-218.
Zhou S, Feng J, Xie Q, et al. Traditional Chinese medicine Shenhuang granule in patients with severe/critical COVID-19:A randomized controlled multicenter trial[J]. Phytomedicine, 2021, 89:153612.
李彦,廉富,杨欣.黄芩苷防治脓毒症相关器官损伤的药理作用研究进展[J]. 现代药物与临床, 2025, 40(11):2960-2964.
刘薇,郑爔,王东,等.基于动物实验、网络药理学结合分子对接技术探讨四磨汤治疗脓毒症的分子机制[J]. 药物评价研究, 2025, 48(3):689-697.
牟娜,卓阳,高敏洁.医疗机构中药人用经验收集现状分析及对策建议[J]. 中国新药与临床杂志, 2024,43(8):582-586.
杨飞宇,郭力恒,张敏州,等.基于“四证四法”辨治脓毒症心肌病的研究进展[J]. 中华中医药杂志, 2022,37(8):4263-4268.
Zhang L, Yang L K, Xie X W, et al. Baicalin magnesium salt attenuates lipopolysaccharide-induced acute lung injury via inhibiting of TLR4/NF-κB signaling pathway[J]. J Immunol Res, 2021, 2021:6629531.
Jiang C, Zhang J C, Xie H W, et al. Baicalein suppresses lipopolysaccharide-induced acute lung injury by regulating Drp1-dependent mitochondrial fission of macrophages[J]. Biomed Pharmacother, 2022, 145:112408.
Lu Q R, Wang N, Wen D F, et al. Baicalin attenuates lipopolysaccharide-induced intestinal inflammatory injury via suppressing PARP1-mediated NF-κB and NLRP3signalling pathway[J]. Toxicon, 2024, 239:107612.
单嘉湘,李淑芳.中医治疗脓毒症凝血功能障碍的研究进展[J]. 中国中医急症, 2023, 32(4):749-752.
王晨燕,孙鑫,陈伟.脓毒症相关凝血功能障碍的中医药研究进展[J]. 中国中医急症, 2023, 32(11):2065-2068.
Wang L L, Feng J, Zhan D Q, et al. Protective effects of tanshinone IIA on sepsis-induced multiple organ dysfunction:A literature review[J]. J Tradit Chin Med,2023, 43(5):1040-1046.
张晓慧,陈聪敏,丁新耘,等.丹参酮IIA通过调控中性粒细胞胞外诱捕网保护脓毒症小鼠急性肺损伤[J]. 陆军军医大学学报, 2023, 45(1):38-44.
夏召弟,马若昕,王凤云,等.党参免疫调节活性及其机制的研究进展[J]. 中草药, 2023, 54(13):4334-4345.
Park J, Yasir M, Han E T, et al. Exploring the therapeutic potential of Rehmannia glutinosa:A network pharmacology and molecular docking analysis across multiple diseases[J]. Curr Issues Mol Biol, 2025, 47(5):329.
Tian W C, Song P, Zang J H, et al. Tanshinone IIA, a component of Salvia miltiorrhiza Bunge, attenuated sepsis-induced liver injury via the SIRT1/Sestrin2/HO-1signaling pathway[J]. J Ethnopharmacol, 2025, 340:119169.
曹强,郭亚菲,叶蕾蕾,等.大黄及其活性成分抗炎作用及机制的研究进展[J]. 中草药, 2022, 53(22):7285-7294.
程记伟,戴璇,陈少裕,等.大黄基于肠道菌群调节治疗脓毒症的研究进展[J]. 中国医药导报, 2023, 20(36):41-45.
王宝恩,黄光伟,任爱民,等.大黄对脓毒症大鼠肠道菌群紊乱影响的实验研究[J]. 临床和实验医学杂志,2012, 11(11):897-898.
Tian Y X, Shen X R, Hu T T, et al. Structural analysis and blood-enriching effects comparison based on biological potency of Angelica sinensis polysaccharides[J]. Front Pharmacol, 2024, 15:1405342.
Zhang M T, Geng W, Guan X Q, et al. Antioxidant and anti-inflammatory effects of different ratios and preparations of Angelica sinensis and Chuanxiong Rhizoma extracts[J]. J Ethnopharmacol, 2024, 334:118559.
Huang L F, Wang J N, Wang L T, et al. Phosphorylated Scrophularia ningpoensis polysaccharides mitigate oxidative stress and modulate immunity by regulating the gut microbiota in cyclophosphamide-treated mice[J]. Food Biosci, 2025, 72:107463.
Chen L P, Li M X, Yang Z Q, et al. Gardenia jasminoides Ellis:Ethnopharmacology, phytochemistry, and pharmacological and industrial applications of an important traditional Chinese medicine[J]. J Ethnopharmacol, 2020, 257:112829.
吕欣悦,张瑞芬,贾婷婷,等.脓毒症胃肠功能障碍中医用药规律的数据挖掘研究[J]. 中国中医急症, 2025,34(4):570-576.
杨思苑,吴铁生,蔺俊杰,等.基于数据挖掘分析中医药治疗脓毒症相关急性肾损伤用药规律[J]. 新中医,2025, 57(17):1-6.
Lu P H, Keng J L, Kuo K L, et al. An apriori algorithmbased association rule analysis to identify herb combinations for treating uremic pruritus using Chinese herbal bath therapy[J]. Evid Based Complement Alternat Med, 2020, 2020:8854772.
Chua M T, Boon Y, Lee Z Y, et al. The role of artificial intelligence in sepsis in the emergency department:A narrative review[J]. Ann Transl Med, 2025, 13(1):4.
Mușat F, Păduraru D N, Bolocan A, et al. Machine learning models in sepsis outcome prediction for ICU patients:Integrating routine laboratory tests-a systematic review[J]. Biomedicines, 2024, 12(12):2892.

2026年第57卷第1期

PDF下载

引用本文

BibTeX

文章信息

doi: 10.7501/j.issn.0253-2670.2026.01.020

接收时间：2025-10-11
首发时间：2026-09-09

补充材料

相关文章

文章信息

作者

出版历史

收稿日期：2025-10-11

基金

作者信息

参考文献

分享链接

https://castjournals.cast.org.cn/joweb/zcy/CN/10.7501/j.issn.0253-2670.2026.01.020

分享至

全文二维码

扫描看全文

引用本文

BibTeX

本文的引用情况

2种不同金属材料的力学参数

科 Family	属数 Number of genus	种数 Number of species	占总种数比例 Percentage of total species (%)	属 Genus	种数 Number of species	占总种数比例 Percentage of total species (%)
鹅膏菌科Amanitaceae	2	11	5.26	鹅膏菌属 Amanita	10	4.78
小菇科 Mycenaceae	2	12	5.74	丝盖伞属 Inocybe	5	2.39
多孔菌科 Polyporaceae	8	14	6.70	蜡蘑属 Laccaria	5	2.39
红菇科 Russulaceae	3	23	11.00	小皮伞属 Marasmius	6	2.87
				小菇属 Mycena	11	5.26
				光柄菇属 Pluteus	5	2.39
				红菇属 Russula	17	8.13
				栓菌属 Trametes	5	2.39

关闭全屏

BibTeX
EndNote
RefWorks
TxT